Define relationships among bone mineral content (BMC), calcium intake and estrogen levels in women runners who run at least 25 miles/week. Ninety eumenorrheic, oligomenorrheic and amenorrheic women as well as thirty eumenorrheic non-runners will be followed for three years. One half of the women will be randomly assigned to a calcium supplementation group while the remaining subjects will be assigned to a placebo group.